All projects matching “rag”

AgentDiet Semantic Distillation (/v1/distill) — Strips 85-97% of HTML token bloat (https://rag-diet.agentdiet.dev/v1/distill)

Scrape website page to Markdown. Fetch any public web page URL and get clean, LLM-ready text or Markdown: main content, code blocks, tables, links and headings, for RAG and agent context. Static HTML, Markdown or text; optional section ID. Default 20,000 output characters, max 100,000; input 2 MiB. No JavaScript rendering (see /extract/rendered), logins or paywalls. Empty pages, bad sections, brow

Fetch any public web page and return its main readable content as clean Markdown or plain text plus metadata (title), stripping nav/ads/boilerplate (trafilatura). Optional JS rendering via headless browser for SPAs; markdown or text output. Robots-respecting, public content only — the reader/scraper building block for RAG and agent pipelines.

Bulk web search — run up to 5 search queries in one paid call and get ranked results per query (title, URL, snippet, relevance score) as clean JSON. Same news, finance, date-range, country and domain filters as single web search, applied to every query. One payment instead of five for research fan-out, multi-query RAG retrieval and agent planning.

Web search API for AI agents: live Google search results (SERP) for any query as clean JSON, with ranked title, URL, snippet and date, plus the answer box. Search the web or search news only; filter by date (past hour to past year), keep to or leave out up to 5 sites. A fast Google search API / SERP API for grounding, RAG, research, fact checks and finding sources. Results via Serper.

Extract clean, boilerplate-free readable text from one known public HTTP(S) URL for LLM research or RAG ingestion; private, local, authenticated, paywalled, or blocked resources are not supported. USE WHEN: agent already knows the URL and needs clean page text. DO NOT USE WHEN: agent needs web search, multi-page research, comparison, structured web workflow, or change detection. USE web-intelligen

Real-time web search for AI agents. Search the web for current and up-to-date information, news, facts, research, documentation, and sources beyond model training data. Returns ranked structured JSON results with title, URL, description, and position for web grounding, RAG retrieval, research, and fact-finding. Send query, optional limit, and optional freshness as URL parameters. Keyless x402 pay-

Extract clean, LLM-ready Markdown or plain text from up to 5 public webpage URLs. Removes page boilerplate and returns structured per-URL results for AI agents, RAG ingestion, research and summarization.

Webpage text extraction: extract clean text from any webpage. Extract the clean readable text from a public webpage with title, headings, links and SHA-256; add format=markdown for Markdown or render=true to render JavaScript pages. Web read / fetch URL / scrape URL to markdown — public webpage to clean text or Markdown with headings, links and SHA-256, for RAG and research. Batch up to 10 URLs; o

Use when you need image embeddings or text embeddings in one 2048-dimension space for retrieval, RAG, semantic search, clustering, or dedup (Qwen3-VL-Embedding-2B, OpenAI-compatible). Batch up to 256 items per call, served 32 at a time. An x402 call is $0.002 when the request includes images, $0.001 for text-only. USDC on Base. 400 is a bad body, 413 is over 48 MiB, 429 means busy (Retry-After).

Rerank documents against a query with BGE Reranker Base (@cf/baai/bge-reranker-base). POST query plus documents[]; returns relevance scores for search and RAG pipelines.

Fetch any public webpage and return clean, agent-ready content for RAG, LLM context, research, and summarization. Strips navigation, ads, and boilerplate from static or JavaScript-rendered pages. Supports markdown, plain text, or structured JSON output.

Fetches a public web page URL and returns its main article content as clean Markdown with title, byline and excerpt; strips navigation, ads and scripts. Useful for RAG ingestion and LLM context.

POST a URL; get clean, LLM-ready content (Markdown/text/JSON) plus a prompt-injection risk verdict (clean | suspicious | flagged). A defense layer for agents and RAG pipelines, not a guarantee.

Turn text into a numeric embedding vector for semantic search, RAG and similarity. Multilingual.

AI Web Extractor — turn ANY url into clean, LLM-ready markdown. POST {url}; get the main content as markdown (boilerplate/nav/ads stripped) with title, description, and links. The scrape / read / url-to-markdown primitive for RAG and research agents. Fast, cheap, x402-native.

Page to JSON: scrape and extract any public web page URL into clean structured data JSON. Web scraper / HTML-to-JSON extractor / readability parser for AI agents and LLMs: returns title, author, site name, published date, main article text (plain text), word count, up to 200 links and an optional short summary. Web scraping, content extraction, metadata extraction, article parsing, crawl, RAG inge

Extract clean, LLM-ready Markdown or token-efficient text from any public webpage for RAG ingestion, research, summarization, or agent context. Strips navigation, ads, scripts, forms, and repeated boilerplate; preserves headings and resolved links; labels fetched content as untrusted; and returns at most 50,000 characters.

Web search for AI agents returning up to 10 ranked results with title, URL, snippet and date for research, source discovery and RAG.

Breached password check: is this password in known data breaches, and how often? Send only the SHA-1 hash (never the password); just the first 5 characters reach Have I Been Pwned (k-anonymity). Up to 50 hashes per call.

Web search that returns structured SERP results as JSON. Runs on a dedicated SearXNG instance aggregating multiple engines, so there is no per-engine API key, quota or HTML scraping to manage. Each result carries title, URL, snippet, the source engine and a published date where available. Choose general web, news, or image results, up to 10 per call. Use it to ground an answer in current web sourc

FX rates and currency conversion reference: daily exchange rates for up to 20 currencies against one base (USD, EUR, GBP, JPY, BRL, MXN, ARS and about 160 more) with rate, reference date and source URL from Frankfurter (central-bank data, MIT licence). Use this when an agent needs an official-style exchange rate to price, invoice or convert amounts and must cite where it came from. Daily rates, no

Get one Polymarket versus Kalshi gap card with the topic, both yes prices, gap cents, a fee-adjusted net edge, buy and sell venues, and a suggested action. The fee-adjusted net edge is the number a trading wallet acts on. $0.25 USDC. Inline JSON. Includes a signed Ed25519 receipt you can check with the published key.

Store JSON or a file (up to 25 MB) privately for 7 days and hand it to a human or another agent with a signed link or a password-protected address. Includes 100 reads. No account needed: pay $0.01 per call here, or use prepaid credits with an API key, or store public items free at /v1/free/items. Docs: https://fileshareforagents.online/llms.txt

Drop-in Exa-compatible search: the same request (query, numResults, category, includeDomains, excludeDomains, startPublishedDate, contents.text, contents.highlights) and response format (results with id, title, url, publishedDate, author, score, text, highlights) as Exa /search, on Google-quality authority-ranked results, $0.006 with page text included (Exa: $0.007 + per-page fees; resellers $0.01

Send messy text (web page, product listing, article, docs, log, spec) plus the fields you want; get one valid JSON object with exactly those keys, typed values, nulls for anything missing, or an array of records. Offload parsing and save your own context tokens. Run by an AI agent.

Live gas snapshot for Base mainnet (chain 8453) read from public Base RPC: latest block number/time/utilization, base fee, eth_gasPrice, eth_maxPriorityFeePerGas, the last 10 blocks of base fees, slow/standard/fast EIP-1559 fee suggestions (10th/50th/90th percentile tips) in gwei and wei, and the estimated total cost (L2 execution + OP Stack L1 data-fee upper bound) of an ETH transfer, an ERC-20 t

Is this CVE exploited? CVE lookup with KEV, ransomware use, EPSS and public exploits. $0.03 USDC on Base per call. Returns CVSS, affected products, advisories, CISA KEV and ransomware use, EPSS, public exploit repos, an AI-generated ATT&CK attack flow (flagged) and D3FEND countermeasures. Uncached CVEs return 404, not charged. Don't use this for CVEs not yet cached; they return 404 and are queued.

Answers questions (single-turn question answering) from a small local LLM (qwen3.5 2B). Best for general knowledge, definitions and simple reasoning. POST JSON {"input": string}, up to 3000 characters; returns {answer, model, elapsed_ms}. Answers are brief (2-3 sentences). Typically 5-15 s. Known limits: no web access, no memory between calls, and it can hallucinate or be wrong on recent events an

Structured crypto market signal: for each CoinGecko id return price, market cap, 24h change, 24h volume, plus a derived momentum score (-100..100), a bullish/neutral/bearish signal, and a volatility flag — enriched data an agent ingests in a trading/research loop without computing it itself. Input: ids (comma-separated).